Birds demonstrate that flapping-wing flight (FWF) is a versatile flight mode, compatible with hov... more Birds demonstrate that flapping-wing flight (FWF) is a versatile flight mode, compatible with hovering, forward flight and gliding to save energy. This extended flight domain would be especially useful on mini-UAVs. However, design is challenging because aerodynamic efficiency is conditioned by complex movements of the wings, and because many interactions exist between morphological (wing area, aspect ratio) and kinematic parameters (flapping frequency, stroke amplitude, wing unfolding). Here we used artificial evolution to optimize these morpho-kinematic features on a simulated 1 kg UAV, equipped with wings articulated at the shoulder and wrist. Flight tests were conducted in a dedicated steady aerodynamics simulator. Parameters generating horizontal flight for minimal mechanical power were retained. Results showed that flight at medium speed (10-12 m s −1 ) can be obtained for reasonable mechanical power (20 W kg −1 ), while flight at higher speed (16-20 m s −1 ) implied increased power (30-50 W kg −1 ). Flight at low speed (6-8 m s −1 ) necessitated unrealistic power levels (70-500 W kg −1 ), probably because our simulator neglected unsteady aerodynamics. The underlying adaptation of morphology and kinematics to varying flight speed were compared to available biological data on the flight of birds.
Abstract Birds demonstrate that flapping-wing flight (FWF) is a versatile flight mode, compatible... more Abstract Birds demonstrate that flapping-wing flight (FWF) is a versatile flight mode, compatible with hovering, forward flight and gliding to save energy. This extended flight domain would be especially useful on mini-UAVs. However, design is challenging because aerodynamic efficiency is conditioned by complex movements of the wings, and because many interactions exist between morphological (wing area, aspect ratio) and kinematic parameters (flapping frequency, stroke amplitude, wing unfolding).
Abstract In this study we develop a feedback controller for a four wheeled autonomous mobile robo... more Abstract In this study we develop a feedback controller for a four wheeled autonomous mobile robot. The purpose of the controller is to guarantee robust performance of an aggressive maneuver (90 degrees turn) at high velocity (about 10 m/s) on a loose surface (dirty road). To tackle this highly nonlinear control problem, we employ multi-objective evolutionary algorithms to explore and optimize the parameters of a neural network-based controller.
ABSTRACT The present paper analyzes the mutual relationships between generative and developmental... more ABSTRACT The present paper analyzes the mutual relationships between generative and developmental systems (GDS) and synaptic plasticity when evolving plastic artificial neural networks (ANNs) in reward-based scenarios. We first introduce the concept of synaptic Transitive Learning Abilities (sTLA), which reflects how well an evolved plastic ANN can cope with learning scenarios not encountered during the evolution process.
Abstract Developmental robotics (also known as epigenetic robotics) is mainly concerned with mode... more Abstract Developmental robotics (also known as epigenetic robotics) is mainly concerned with modeling the postnatal development of cognitive behaviors in living systems, such as language, emotion, curiosity, anticipation, and social skills. While current work in this field has shown significant successes, we believe integrating research on developmental (including epigenetic and morphogenetic) robotics and evolutionary robotics is the natural next step.
Abstract—This paper introduces a hierarchy of concepts to classify the goals and the methods of w... more Abstract—This paper introduces a hierarchy of concepts to classify the goals and the methods of works that mix neuroevolution and synaptic plasticity. We propose definitions of “behavioral robustness” and oppose it to “reward-based behavioral changes”; we then distinguish the switch between behaviors and the acquisition of new behaviors. Last, we summarize the concept of “synaptic General Learning Abilities”(sGLA) and that of “synaptic Transitive learning Abilities (sTLA).
Abstract The reality gap, that often makes controllers evolved in simulation inefficient once tra... more Abstract The reality gap, that often makes controllers evolved in simulation inefficient once transferred onto the physical robot, remains a critical issue in Evolutionary Robotics (ER). We hypothesize that this gap highlights a conflict between the efficiency of the solutions in simulation and their transferability from simulation to reality: the most efficient solutions in simulation often exploit badly modeled phenomena to achieve high fitness values with unrealistic behaviors.
Abstract: Damage recovery is critical for autonomous robots that need to operate for a long time ... more Abstract: Damage recovery is critical for autonomous robots that need to operate for a long time without assistance. Most current methods are complex and costly because they require anticipating each potential damage in order to have a contingency plan ready. As an alternative, we introduce the T-resilience algorithm, a new algorithm that allows robots to quickly and autonomously discover compensatory behaviors in unanticipated situations.
Abstract: A central biological question is how natural organisms are so evolvable (capable of qui... more Abstract: A central biological question is how natural organisms are so evolvable (capable of quickly adapting to new environments). A key driver of evolvability is the widespread modularity of biological networks--their organization as functional, sparsely connected subunits--but there is no consensus regarding why modularity itself evolved.
Wheel-legged hybrid robots promise to combine the efficiency of wheeled robots with the versatili... more Wheel-legged hybrid robots promise to combine the efficiency of wheeled robots with the versatility of legged robots: they are able to roll on simple terrains, to dynamically adapt their posture and even to walk on uneven grounds. Although different locomotion modes of such robots have been studied, a pivotal question remains: how to automatically adapt the locomotion mode when the environment changes? We here propose that the robot autonomously discovers its locomotion mode using optimization-based learning. To that aim, we introduce a new algorithm that relies on a forward model and a stochastic multi-objective optimization. Three objectives are optimized: (1) the average displacement speed, (2) the expended energy and (3) the transferability score, which reflects how well the behavior of the robot is in agreement with the predictions of the forward model. This transferability function is approximated by conducting 20 experiments of one second on the real robot during the optimization. In the three investigated situations (flat ground, grass-like terrain, tunnel-like environment), our method found efficient controllers for forward locomotion in 1 to 2 minutes: the robot used its wheels on the flat ground, it walked on the grass-like terrain and moved with a lowered body in the tunnellike environment.
… in Artificial Life. Darwin Meets von …, Jan 1, 2011
Genetic Regulation Networks (GRNs) are a model of the mechanisms by which a cell regulates the ex... more Genetic Regulation Networks (GRNs) are a model of the mechanisms by which a cell regulates the expression of its different genes depending on its state and the surrounding environment. These mechanisms are thought to greatly improve the capacity of the evolutionary process through the regulation loop they create. Some Evolutionary Algorithms have been designed to offer improved performance by taking advantage of the GRN mechanisms. A recent hypothesis suggests a correlation between the length of promoters for a gene and the complexity of its activation behavior in a given genome. This hypothesis is used to identify the links in in-vivo GRNs in a recent paper and is also interesting for evolutionary algorithms. In this work, we first confirm the correlation between the length of a promoter (binding site) and the complexity of the interactions involved on a simplified model. We then show that an operator modifying the length of the promoter during evolution is useful to converge on complex specific network topologies. We used the Analog Genetic Encoding (AGE) model in order to test our hypothesis.
Proceedings of the 13th annual conference …, Jan 1, 2011
Abstract Wheel-legged hybrid robots are versatile machines that can employ several locomotion mod... more Abstract Wheel-legged hybrid robots are versatile machines that can employ several locomotion modes; however, automatically choosing the right locomotion mode is still an open problem in robotics. We here propose that the robot autonomously discovers its locomotion mode using a multi-objective evolutionary optimization and a fixed internal model. Three objectives are optimized:(1) the displacement speed computed with the internal model,(2) the predicted expended energy and (3) the transferability score, which ...
In Evolutionary Robotics (ER), controllers are assessed in a single or a few environments. As a c... more In Evolutionary Robotics (ER), controllers are assessed in a single or a few environments. As a consequence, good performances in new different contexts are not guaranteed. While a lot of ER works deal with robustness, i.e. the ability to perform well on new contexts close to the ones used for evaluation, no current approach is able to promote broader generalisation abilities without any assumption on the new contexts. In this paper, we introduce the ProGAb approach, which is based on the standard three data sets methodology of supervised machine learning, and compare it to state-ofthe-art ER methods on two simulated robotic tasks: a navigation task in a T-maze and a more complex ball-collecting task in an arena. In both applications, the ProGAb approach: (1) produced controllers with better generalisation abilities than the other methods; (2) needed two to three times fewer evaluations to discover such solutions.
In this study we present a chain sliding mode controller for the control of a four wheeled autono... more In this study we present a chain sliding mode controller for the control of a four wheeled autonomous mobile robot performing aggressive turning maneuver to 90 degrees on a loose surface. The controller consists of a set of local sliding mode controllers and the hyperplanes of switching between them. The parameters of the sliding mode controllers and the hyperplanes are obtained using methods of multiobjective stochastic optimization applied to a model of the robot. The obtained controller is used to drive the mobile robot. The results show that is capable to control the robot during aggressive maneuver. In particular, the steering radius obtained with the controller was two times smaller then the minimal steering radius admitted by the robot.
Many controllers for complex agents have been successfully generated by automatically desiging ar... more Many controllers for complex agents have been successfully generated by automatically desiging artificial neural networks with evolutionary algorithms. However, typical evolved neural networks are not able to adapt themselves online, making them unable to perform tasks that require online adaptation. Nature solved this problem on animals with plastic nervous systems. Inpired by neuroscience models of plastic neuralnetwork, the present contribution proposes to use a combination of Hebbian learning, neuro-modulation and a a generative mapbased encoding. We applied the proposed approach on a problem from operant conditioning (a Skinner box), in which numerous different association rules can be learned. Results show that the map-based encoding scaled up better than a classic direct encoding on this task. Evolving neural networks using a mapbased generative encoding also lead to networks that works with most rule sets even when the evolution is done on a small subset of all the possible cases. Such a generative encoding therefore appears as a key to improve the generalization abilities of evolved adaptive neural networks.
Evolutionary Robotics (ER) aims at automatically designing robots or controllers of robots withou... more Evolutionary Robotics (ER) aims at automatically designing robots or controllers of robots without having to describe their inner workings. To reach this goal, ER researchers primarily employ phenotypes that can lead to an infinite number of robot behaviors and fitness functions that only reward the achievement of the task-and not how to achieve it. These choices make ER particularly prone to premature convergence.
This paper considers the field of Evolutionary Robotics (ER) from the perspective of its potentia... more This paper considers the field of Evolutionary Robotics (ER) from the perspective of its potential users: roboticists. The core hypothesis motivating this field of research is discussed, as well as the potential use of ER in a robot design process. Four main aspects of ER are presented: (a) ER as an automatic parameter tuning procedure, which is the most mature application and is used to solve real robotics problem, (b) evolutionary-aided design, which may benefit the designer as an efficient tool to build robotic systems (c) ER for online adaptation, i.e. continuous adaptation to changing environment or robot features and (d) automatic synthesis, which corresponds to the automatic design of a mechatronic device and its control system. Critical issues are also presented as well as current trends and pespectives in ER. A section is devoted to a roboticist's point of view and the last section discusses the current status of the field and makes some suggestions to increase its maturity. S. Doncieux et al. (Eds.): New Horizons in Evolutionary Robotics, SCI 341, pp. 3-25.
In this work we regard the problem of trajectory planning for aggressive maneuver of a wheeled mo... more In this work we regard the problem of trajectory planning for aggressive maneuver of a wheeled mobile robot on loose surface. Our approach is inspired by previously reported analysis of professional rally racers actions during sharp turn. Using numerical simulations we obtain a set of solutions representing all possible compromises between speed and accuracy of maneuver. We choose a particular solution and extend it by producing a continuous mapping from desired trajectory turn angles to parameters of the control inputs.
Birds demonstrate that flapping-wing flight (FWF) is a versatile flight mode, compatible with hov... more Birds demonstrate that flapping-wing flight (FWF) is a versatile flight mode, compatible with hovering, forward flight and gliding to save energy. This extended flight domain would be especially useful on mini-UAVs. However, design is challenging because aerodynamic efficiency is conditioned by complex movements of the wings, and because many interactions exist between morphological (wing area, aspect ratio) and kinematic parameters (flapping frequency, stroke amplitude, wing unfolding). Here we used artificial evolution to optimize these morpho-kinematic features on a simulated 1 kg UAV, equipped with wings articulated at the shoulder and wrist. Flight tests were conducted in a dedicated steady aerodynamics simulator. Parameters generating horizontal flight for minimal mechanical power were retained. Results showed that flight at medium speed (10-12 m s −1 ) can be obtained for reasonable mechanical power (20 W kg −1 ), while flight at higher speed (16-20 m s −1 ) implied increased power (30-50 W kg −1 ). Flight at low speed (6-8 m s −1 ) necessitated unrealistic power levels (70-500 W kg −1 ), probably because our simulator neglected unsteady aerodynamics. The underlying adaptation of morphology and kinematics to varying flight speed were compared to available biological data on the flight of birds.
Abstract Birds demonstrate that flapping-wing flight (FWF) is a versatile flight mode, compatible... more Abstract Birds demonstrate that flapping-wing flight (FWF) is a versatile flight mode, compatible with hovering, forward flight and gliding to save energy. This extended flight domain would be especially useful on mini-UAVs. However, design is challenging because aerodynamic efficiency is conditioned by complex movements of the wings, and because many interactions exist between morphological (wing area, aspect ratio) and kinematic parameters (flapping frequency, stroke amplitude, wing unfolding).
Abstract In this study we develop a feedback controller for a four wheeled autonomous mobile robo... more Abstract In this study we develop a feedback controller for a four wheeled autonomous mobile robot. The purpose of the controller is to guarantee robust performance of an aggressive maneuver (90 degrees turn) at high velocity (about 10 m/s) on a loose surface (dirty road). To tackle this highly nonlinear control problem, we employ multi-objective evolutionary algorithms to explore and optimize the parameters of a neural network-based controller.
ABSTRACT The present paper analyzes the mutual relationships between generative and developmental... more ABSTRACT The present paper analyzes the mutual relationships between generative and developmental systems (GDS) and synaptic plasticity when evolving plastic artificial neural networks (ANNs) in reward-based scenarios. We first introduce the concept of synaptic Transitive Learning Abilities (sTLA), which reflects how well an evolved plastic ANN can cope with learning scenarios not encountered during the evolution process.
Abstract Developmental robotics (also known as epigenetic robotics) is mainly concerned with mode... more Abstract Developmental robotics (also known as epigenetic robotics) is mainly concerned with modeling the postnatal development of cognitive behaviors in living systems, such as language, emotion, curiosity, anticipation, and social skills. While current work in this field has shown significant successes, we believe integrating research on developmental (including epigenetic and morphogenetic) robotics and evolutionary robotics is the natural next step.
Abstract—This paper introduces a hierarchy of concepts to classify the goals and the methods of w... more Abstract—This paper introduces a hierarchy of concepts to classify the goals and the methods of works that mix neuroevolution and synaptic plasticity. We propose definitions of “behavioral robustness” and oppose it to “reward-based behavioral changes”; we then distinguish the switch between behaviors and the acquisition of new behaviors. Last, we summarize the concept of “synaptic General Learning Abilities”(sGLA) and that of “synaptic Transitive learning Abilities (sTLA).
Abstract The reality gap, that often makes controllers evolved in simulation inefficient once tra... more Abstract The reality gap, that often makes controllers evolved in simulation inefficient once transferred onto the physical robot, remains a critical issue in Evolutionary Robotics (ER). We hypothesize that this gap highlights a conflict between the efficiency of the solutions in simulation and their transferability from simulation to reality: the most efficient solutions in simulation often exploit badly modeled phenomena to achieve high fitness values with unrealistic behaviors.
Abstract: Damage recovery is critical for autonomous robots that need to operate for a long time ... more Abstract: Damage recovery is critical for autonomous robots that need to operate for a long time without assistance. Most current methods are complex and costly because they require anticipating each potential damage in order to have a contingency plan ready. As an alternative, we introduce the T-resilience algorithm, a new algorithm that allows robots to quickly and autonomously discover compensatory behaviors in unanticipated situations.
Abstract: A central biological question is how natural organisms are so evolvable (capable of qui... more Abstract: A central biological question is how natural organisms are so evolvable (capable of quickly adapting to new environments). A key driver of evolvability is the widespread modularity of biological networks--their organization as functional, sparsely connected subunits--but there is no consensus regarding why modularity itself evolved.
Wheel-legged hybrid robots promise to combine the efficiency of wheeled robots with the versatili... more Wheel-legged hybrid robots promise to combine the efficiency of wheeled robots with the versatility of legged robots: they are able to roll on simple terrains, to dynamically adapt their posture and even to walk on uneven grounds. Although different locomotion modes of such robots have been studied, a pivotal question remains: how to automatically adapt the locomotion mode when the environment changes? We here propose that the robot autonomously discovers its locomotion mode using optimization-based learning. To that aim, we introduce a new algorithm that relies on a forward model and a stochastic multi-objective optimization. Three objectives are optimized: (1) the average displacement speed, (2) the expended energy and (3) the transferability score, which reflects how well the behavior of the robot is in agreement with the predictions of the forward model. This transferability function is approximated by conducting 20 experiments of one second on the real robot during the optimization. In the three investigated situations (flat ground, grass-like terrain, tunnel-like environment), our method found efficient controllers for forward locomotion in 1 to 2 minutes: the robot used its wheels on the flat ground, it walked on the grass-like terrain and moved with a lowered body in the tunnellike environment.
… in Artificial Life. Darwin Meets von …, Jan 1, 2011
Genetic Regulation Networks (GRNs) are a model of the mechanisms by which a cell regulates the ex... more Genetic Regulation Networks (GRNs) are a model of the mechanisms by which a cell regulates the expression of its different genes depending on its state and the surrounding environment. These mechanisms are thought to greatly improve the capacity of the evolutionary process through the regulation loop they create. Some Evolutionary Algorithms have been designed to offer improved performance by taking advantage of the GRN mechanisms. A recent hypothesis suggests a correlation between the length of promoters for a gene and the complexity of its activation behavior in a given genome. This hypothesis is used to identify the links in in-vivo GRNs in a recent paper and is also interesting for evolutionary algorithms. In this work, we first confirm the correlation between the length of a promoter (binding site) and the complexity of the interactions involved on a simplified model. We then show that an operator modifying the length of the promoter during evolution is useful to converge on complex specific network topologies. We used the Analog Genetic Encoding (AGE) model in order to test our hypothesis.
Proceedings of the 13th annual conference …, Jan 1, 2011
Abstract Wheel-legged hybrid robots are versatile machines that can employ several locomotion mod... more Abstract Wheel-legged hybrid robots are versatile machines that can employ several locomotion modes; however, automatically choosing the right locomotion mode is still an open problem in robotics. We here propose that the robot autonomously discovers its locomotion mode using a multi-objective evolutionary optimization and a fixed internal model. Three objectives are optimized:(1) the displacement speed computed with the internal model,(2) the predicted expended energy and (3) the transferability score, which ...
In Evolutionary Robotics (ER), controllers are assessed in a single or a few environments. As a c... more In Evolutionary Robotics (ER), controllers are assessed in a single or a few environments. As a consequence, good performances in new different contexts are not guaranteed. While a lot of ER works deal with robustness, i.e. the ability to perform well on new contexts close to the ones used for evaluation, no current approach is able to promote broader generalisation abilities without any assumption on the new contexts. In this paper, we introduce the ProGAb approach, which is based on the standard three data sets methodology of supervised machine learning, and compare it to state-ofthe-art ER methods on two simulated robotic tasks: a navigation task in a T-maze and a more complex ball-collecting task in an arena. In both applications, the ProGAb approach: (1) produced controllers with better generalisation abilities than the other methods; (2) needed two to three times fewer evaluations to discover such solutions.
In this study we present a chain sliding mode controller for the control of a four wheeled autono... more In this study we present a chain sliding mode controller for the control of a four wheeled autonomous mobile robot performing aggressive turning maneuver to 90 degrees on a loose surface. The controller consists of a set of local sliding mode controllers and the hyperplanes of switching between them. The parameters of the sliding mode controllers and the hyperplanes are obtained using methods of multiobjective stochastic optimization applied to a model of the robot. The obtained controller is used to drive the mobile robot. The results show that is capable to control the robot during aggressive maneuver. In particular, the steering radius obtained with the controller was two times smaller then the minimal steering radius admitted by the robot.
Many controllers for complex agents have been successfully generated by automatically desiging ar... more Many controllers for complex agents have been successfully generated by automatically desiging artificial neural networks with evolutionary algorithms. However, typical evolved neural networks are not able to adapt themselves online, making them unable to perform tasks that require online adaptation. Nature solved this problem on animals with plastic nervous systems. Inpired by neuroscience models of plastic neuralnetwork, the present contribution proposes to use a combination of Hebbian learning, neuro-modulation and a a generative mapbased encoding. We applied the proposed approach on a problem from operant conditioning (a Skinner box), in which numerous different association rules can be learned. Results show that the map-based encoding scaled up better than a classic direct encoding on this task. Evolving neural networks using a mapbased generative encoding also lead to networks that works with most rule sets even when the evolution is done on a small subset of all the possible cases. Such a generative encoding therefore appears as a key to improve the generalization abilities of evolved adaptive neural networks.
Evolutionary Robotics (ER) aims at automatically designing robots or controllers of robots withou... more Evolutionary Robotics (ER) aims at automatically designing robots or controllers of robots without having to describe their inner workings. To reach this goal, ER researchers primarily employ phenotypes that can lead to an infinite number of robot behaviors and fitness functions that only reward the achievement of the task-and not how to achieve it. These choices make ER particularly prone to premature convergence.
This paper considers the field of Evolutionary Robotics (ER) from the perspective of its potentia... more This paper considers the field of Evolutionary Robotics (ER) from the perspective of its potential users: roboticists. The core hypothesis motivating this field of research is discussed, as well as the potential use of ER in a robot design process. Four main aspects of ER are presented: (a) ER as an automatic parameter tuning procedure, which is the most mature application and is used to solve real robotics problem, (b) evolutionary-aided design, which may benefit the designer as an efficient tool to build robotic systems (c) ER for online adaptation, i.e. continuous adaptation to changing environment or robot features and (d) automatic synthesis, which corresponds to the automatic design of a mechatronic device and its control system. Critical issues are also presented as well as current trends and pespectives in ER. A section is devoted to a roboticist's point of view and the last section discusses the current status of the field and makes some suggestions to increase its maturity. S. Doncieux et al. (Eds.): New Horizons in Evolutionary Robotics, SCI 341, pp. 3-25.
In this work we regard the problem of trajectory planning for aggressive maneuver of a wheeled mo... more In this work we regard the problem of trajectory planning for aggressive maneuver of a wheeled mobile robot on loose surface. Our approach is inspired by previously reported analysis of professional rally racers actions during sharp turn. Using numerical simulations we obtain a set of solutions representing all possible compromises between speed and accuracy of maneuver. We choose a particular solution and extend it by producing a continuous mapping from desired trajectory turn angles to parameters of the control inputs.
Uploads
Papers by J.-B. Mouret